This protocol is testing the hypothesis that microphone site influences the noise dose computed by a dosimeter, otherwise known as an integrating sound level meter. An ancillary goal is to examine individual differences in projected dose measured in the ear canals of the subjects. Three measurement conditions of 20 minutes each are being used. They are as follows: placement of a probe tube microphone into the ear canal a known distance from the eardrum; placement of the microphone on the right shoulder; and placement of the microphone on the left breast pocket. The subject is instructed to resume his/her normal work routine while the noise dose is being monitored. In addition, one stationary measurement is taken with the microphone mounted on a tripod. Data are then downloaded to a PC and analyzed for differences in variance for the different measurements. The study is being conducted with the cooperation of the Division of Safety at the NIH air conditioning chiller plant using NIH employees. A total of 10 adult subjects will be used. To date data have been collected on two subjects. Major differences are shown with higher projected dose being measured at the eardrum vs the other torso positions and the stationary measurement. While not surprising, these data demonstrate the importance of taking into account the acoustics of the entire sound path when assessing risk of hearing loss. These data challenge the conventional wisdom of maintaining homogeneity of measurements by using one microphone site only.